AI ToolsWritingLinkedInProduct

Voice Confidence Explained: What 0-100% Actually Means

Four slices - Brief Sparky, transcripts, approvals, and Post Analytics - combine into one score. Why transcript credit is now flat +2% and what to do if your number dropped.

Voice Confidence Explained: What 0-100% Actually Means
On this page

Voice Confidence is the percentage in the top bar (Voice Profile drawer). It measures how much signal Sparky has learned about how you write - not how good your LinkedIn posts are.

Four slices, one total

The score combines four sources, each with its own cap. Together they can reach 99% - Sparky is always learning, so the meter never quite hits 100%.

  • Brief Sparky (max 10%): strategic context you enter manually - who you are, who you speak to, what you care about
  • Transcripts (max 40%): +2% per distinct processed recording (20 transcripts to max)
  • Your posts (max 40%): approving drafts as-is or after edits, with diminishing returns per approval
  • Analytics (max 5%): +0.5% per published post with synced LinkedIn performance (10 tracked posts to max)

Why transcript credit changed

Transcript credit used to front-load early projects. It is now flat: every distinct transcript earns the same +2%, up to 40%. If you built a high score mostly from volume, your number may look lower after recalibration - your voice profile data is still there. The meter is just honest about how much each source contributes.

The new analytics slice rewards posts that actually performed on LinkedIn, not only posts you approved in the editor. Publish through SparkVox and let Post Analytics sync.

What the score does not mean

  • It is not a quality grade on your LinkedIn posts
  • Re-processing the same project does not add transcript credit again
  • It does not come from LinkedIn scraping - transcripts, brief, approvals, and analytics only

Edits, discards, and regression detection

Approvals grow the Your posts slice. When you edit before approving, SparkVox can extract style corrections from the diff. Discarding a draft also teaches Sparky which topics and post styles to deprioritize on future projects - not only how to phrase the next line. See SparkVox learns from discarded drafts, not just edits.

Voice Health may cluster repeated edits and auto-apply rules (for example stop opening with a question). The Draft Feedback tab and Analytics theme map in the same drawer show proof, corrections, and publish vs edit vs discard patterns by topic. See Voice Health panel.

Project Presets vs voice profile

Voice profile is how you speak from recordings. Project Presets (Settings → Project Presets) is who you are and what you are trying to accomplish. Both shape every post - confidence only measures the four slices above.

Full detail: How voice learning works. Fewer but stronger drafts per project: why SparkVox curates fewer posts per project.

SparkVox

Your recordings are already the content.

Upload a call, podcast, or interview. SparkVox extracts moments and drafts up to 15 LinkedIn posts in your voice. You review every draft before anything publishes.

Start 14-day Pro trial →

Free to sign up. Card required for full access.
No charge until day 15.

You might also like

Your recordings are already the content.

Your calls and interviews are already the content. SparkVox writes the posts in your voice - upload one recording and see the drafts before you're charged.

Start 14-day Pro trial →

Free to sign up. Card required for full access. No charge until day 15.